Power Network Fault Diagnosis Based on Multi - source Network Message Data

Abstract

In the event of a grid failure, the network message data (SMV, GOOSE) contains all the information about the primary and secondary devices. The amount of information data is large, and it is difficult for the personnel to make a quick and accurate judgment about the fault type from the massive data, affecting fault handling and safe grid operation. In this paper, modeling is unified based on network message data about the primary equipment. The comprehensive data analysis between multiple substations can be achieved by acquiring alarm information, network message information and wave recording information through network message analysis devices; meanwhile, the longitudinal differential principle is applied to fault diagnosis, which can realize accurate fault diagnosis of power grid, and provide data basis for accident treatment.